A cognitive analysis of similes in the book of Hosea

(1)

A Cognitive Analysis of Similes

in the Book of Hosea

by

J N POHLIG

Dissertation presented for the degree of

DOCTOR OF LITERATURE in

BIBLICAL LANGUAGES at the

University of Stellenbosch

Promotor: Prof C H J van der Merwe Date submitted: April 2006

(2)

Declaration

I, the undersigned, hereby declare that the work contained in this dissertation is my own original work and has not previously in its entirety or in part been submitted to any university for a degree.

Signature:

(3)

ABSTRACT

This study accounts for the forms and functions of the similes in the Book of Hosea. It proposes new tools for textual criticism, biblical interpretation, and understanding Biblical Hebrew (BH) worldview.

Chapter One presents the task we have chosen for ourselves, its nature, some obstacles from other areas of scholarship, and the foundational notions of embodiment and Prototype Theory.

Chapter Two presents principles drawn from Cognitive Semantics and Cognitive Syntax. A weakened version of the Lakoff-Johnson conceptual metaphor theory is adopted, and the key notions of embodiment and judgments of prototypicality are presented. Elements of Conceptual Blending are presented and adapted for simile analysis. Finally, text-based differences between metaphors and similes are discussed.

Chapter Three presents cognitive cultural constructs of Strauss and Quinn: cultural schemas, cultural exemplars, cultural models, and cultural themes. Strauss and Quinn’s conclusions about metaphors’ use in everyday speech are shown to agree with our postulation of speaker assessment of the hearer’s ability to process utterances before they are produced. This postulation allows us to erect one part of a theory of simile.

Chapter Three then integrates metaphor with the Strauss-Quinn cultural meaning model, and then with Boroditsky’s Weak Structuring view of metaphor. The effect is to provide a reasonable basis, amenable to empirical investigation, for the investigation of both metaphor and simile. Finally, the notions of embodiment and prototypicality are applied to the Strauss-Quinn model.

Chapter Four presents various assumptions and conclusions that are later used to analyze Hosea’s similes. These include: (1) elements of Floor’s (2004a) model of Information Structure for BH narrative, with modifications and additions for poetry; (2) three cognitive types of similes in Hosea, posited for working purposes; (3) an adaptation of the conceptual blending apparatus to similes; (4) hypotheses to account for the distribution of similes versus that of metaphors in BH poetry, and to account for patterned differences in how various kinds of concepts are combined and manipulated; (5) an integration of these patterns with the three simile types; and (6) correlation of the cultural constructs of cultural schema, cultural theme, and cultural model with Hosea’s similes and metaphors.

(4)

Chapter Five presents a number of scholarly views of the Book of Hosea, and characterizes the principal authorities cited in the next chapter.

Chapter Six deductively applies all the foregoing theory to an examination of Hosea’s similes. Other observations are made inductively: (1) kinaesthetic image schemas’ role in Hosea’s poetry; (2) systematic difference in the use of similes versus metaphors in image elaboration; and (3) Information Structure’s role in simile analysis.

Chapter Seven summarizes this study’s research and conclusions concerning, e.g., (1) the criteria for accounting for the embodiment and judgments of prototypicality characterizing Hosea’s similes; (2) the dependence of Hosea and his audience upon knowledge of themselves and their environment for their view of YHWH; and (3) the aid brought by a cognitive theory of similes in the task of textual criticism.

Chapter Eight discusses prospects for further research and possible implications for translating Hosea’s similes and metaphors.

OPSOMMING

Hierdie studie gee rekenskap van die vorme en funksies van die vergelykings in Hosea. Nuwe instrumente vir tekskritiek, bybelse interpretasie en die verstaan van die Bybelse Hebreeuse (BH) wêreldbeeld word hierin voorgestel.

Hoofstuk Een spel die essensie van hierdie ondersoek uit, die aard daarvan, ‘n aantal struiklelblokke vanuit ander vakkundige areas, asook die fundamentele begrippe van beliggaming (“embodiment”) en prototipikaliteit.

Hoofstuk Twee formuleer die beginsels wat uit kognitiewe semantiek en kognitiewe sintaksis ontleen is. ‘n Verswakde vorm van die Lakoff-Johnson konseptuele metafoorteorie word oorgeneem, en die sleutelbegrippe van beliggaming en oordele van prototipikaliteit word verduidelik. Elemente van konseptuele vermenging (“conceptual blending”) word aan die orde gestel en aangepas vir die vergelykingsanalise. Laastens word teksgebaseerde verskille tussen metafore en vergelykings bespreek.

(5)

Hoofstuk Drie stel die kognitiewe kulturele konstrukte van Strauss en Quinn aan die orde: kulturele skemas, kulturele voorbeelde, kulturele modelle, en kulturele temas. Daar word aangetoon dat Strauss en Quinn se gevolgtekkings oor die gebruik van metafore in alledaagse taal, ooreenstem met ons postulasie oor sprekers se evaluering van hoorders se vermoë om uitsprake te verwerk nog voordat dit geproduseer is. Hierdie postulasie stel ons in staat om een gedeelte van die vergelykingsteorie te formuleer. Hoofstuk Drie integreer verder die interpretasie van metafore, eers met die Strauss-Quinn se kulturele betekenismodel, en dan met Boroditsky se swak struktureringsbeskouing van metafoor. Die doel is om ‘n redelike basis, verantwoordbaar aan empiriese navorsing, vir die ondersoek van beide metafoor en vergelyking daar te stel. Laastens word die begrippe van beliggaming en prototipikaliteit by die Strauss-Quinn model aangepas. In Hoofstuk Vier word die volgende aannames en gevolgtrekkings wat gebruik word om Hosea se vergelykings te analiseer beskryf: (1) elemente van Floor (2004a) se model van informasiestruktuur vir BH narratiewe, met aanpassings vir die analisie van poësie; (2) drie kognitiewe tipes vergelykings in Hosea wat as werkshipoteses gebruik word; (3) ‘n aanpassing van die konseptuele vermengings apparaat tot vergelykings; (4) hipoteses om vir die verspreiding van vergelykings in teenstelling met metafore in BH poësie verantwoording te doen, asook vir die verskillende patrone in hoe verskeie tipes konsepte gekombineer en gemanipuleer word; (5) ‘n integrasie van hierdie patrone met die drie vergelykingstipes; en (6) ‘n korrelasie van die kulturele konstrukte van kulturele skema, kulturele tema, en kulturele model met Hosea se vergelykings en metafore.

Hoofstuk Vyf stel ‘n aantal wetenskaplike beskouings oor die Boek van Hosea aan die orde. Besondere aandag word gewy aan geleerdes wie se standpunte in Hoofstuk Ses te berde gebring word.

Hoofstuk Ses pas die voorafgaande teoriese model deduktief aan ‘n ondersoek van Hosea se vergelykings toe. Ander waarnemings word induktief gemaak: (1) kinestetiese beeldskemas se rol in Hosea se poësie; (2) sistematiese verskille in die gebruik van vergelykings teenoor metafore in die verruiming van beelde; en (3) informasiestruktuur se rol in vergelykingsanalise.

Hoofstuk Sewe som hierdie studie se bevindinge ten opsigte van die volgende op: (1) die kriteria vir verantwoording van die beliggaming, asook oordele van prototipikaliteit, wat Hosea se vergelykings karakteriseer; (2) die afhanklikheid van Hosea en sy gehoor van kennis van hulleself en hulle omgewing vir hulle beskouing van JHWH; en (3) die hulp wat deur ‘n kognitiewe teorie van vergelykings toegevoeg is in die taak van tekskritiek.

Hoofstuk Agt bespreek die moontlikhede vir verdere navorsing en moontlike implikasies daarvan vir die vertaling van Hosea se vergelykings en metafore.

(6)

Abbreviations

ANE Ancient Near East, Ancient Near Eastern

BDB The Brown-Driver-Briggs Hebrew and English Lexicon.

BH Biblical Hebrew

HAO Humanization, Animalization, Objectification InfStr Information Structure

INTENS Intensifier

KB Koehler, L. and Baumgartner, W. The Hebrew and Aramaic Lexicon of the Old Testament LXX The Septuagint MT Masoretic Text NP Noun phrase P Plural PP Prepositional phrase S Singular T Tenor

TWOT Theological Wordbook of the Old Testament

V Vehicle

Vb Verb

v. Verse

(12)

Acknowledgements

In connection with this study, I owe much to many. Besides the linguists, lexicographers, commentators, and theorists upon whom it rests—truly we stand on the shoulders of giants—there are those whose personal encouragement and logistical and scholarly help have been crucial to me. Prof. C.H.J. van der Merwe has been always ready with friendship, hope, advice, warnings, and vision. On a number of occasions, a simple question from him opened up entirely new lines of thought. To him go my warmest thanks.

In the course of this study I lost my first wife, Ann H. Pohlig, to cancer. But she always stood by my side with a listening ear and a spirit that gladly endured my absences from her and the children, and with a heart that steadfastly wished and prayed for my success. Without these qualities in her, I could never have embarked on this journey of research and writing.

Within several weeks of her passing away, cancer also claimed the life of Dr. William C. Mann, my good friend and mentor. To him I owe many insights, emphases, and much of whatever wholesome scholarly temperament I possess. He was very versatile in his academic pursuits, and they followed a coherent progression, continuing even now to bring benefits to countless scholars. I am among many who keenly feel his loss.1

I must also acknowledge the friendship and interest of various colleagues in S.I.L. International. Chief among them stands Robert G. Carter, who was always ready to listen to my ideas about Biblical Hebrew.

Late in the course of this study, I found and married my present wife Laura. Thus this project, connected to conceptual blending in Biblical Hebrew, was paralleled by a project of blending two adults and their respective children. I am greatly indebted to Laura for her whole-hearted and interested support of this project. Her own experience as a commercial translator of texts Spanish-to-English has afforded her a solid basis for understanding many of the issues confronted in this study; I am very grateful to her.

1

(13)

My four children, Sarah, Adam, Helen Marie, and Charlotte have been very tolerant of my preoccupation when at home and of my absences from home, even when understanding, sometimes only very imperfectly, the reason for them (Adam remarked once to his mother after I had discussed at length an issue arising from my research, “It is very boring, having a linguist for a father”). On the other hand, Sarah is now enjoying an Introduction to Linguistics course at university, so there is hope that some of my interests have rubbed off onto her. To all my children go my very deep thanks and love.

It has been a great privilege to have worked with God’s Word in this way and to have been aware of his aid in this project. If the result helps anyone in understanding or translating his Word, my efforts will have been amply repaid. If any are able to build upon this study for further understanding, or to react to it so as to bring greater light, so much the better.

(14)

(15)

Chapter One

THE TASK AND ITS THEORETICAL BACKGROUND

1.1 Establishing the task

This study aims to develop a framework within Cognitive Linguistics for discovering how similes work in the Hebrew Bible, the Book of Hosea being examined for this purpose. Besides a concern for the advancement of knowledge for its own sake, this study also has in view the needs of biblical interpretation, to which is linked Bible translation, for which there is a ceaseless imperative to better understand the functions of figures of speech and to develop better ways of translating them. Within the subject of biblical interpretation there is, of course, a large variety of needs, ranging from the need to establish the Hebrew text of the Hebrew Bible, the need to determine its inner discourse-unit boundaries, the need to determine ever more solidly the referential sense of the text, to the need to determine its relevance for its intended audiences.

The phrase “how similes work” could be understood in several ways. One could take the phrase to refer to the distribution of similes in discourse (written discourse, in the case of Hosea2) vis-à-vis other language features, such as metaphors, other tropes, and more literal language; or to the distribution of similes vis discourse structures, such as discourse-unit boundaries, and vis-à-vis discourse notions, such as introduction, development, peak, and so forth. Again, one could understand the phrase “how similes work” to refer to similes’ inner mechanisms: the variety of forms exhibited in Hosea’s similes, how his similes combine various concepts, the word order that characterizes his similes, his similes’ inner logic, and his similes’ communicative goals on a conceptual level. Thirdly, one could take “how similes work” to refer to any possible preferences they might display as to the kind of conceptual combinations they effect.

This study will consider all of the understandings given above of “how similes work.” It will proceed along cognitive lines, so we shall be asking questions such as: Are some similes more “simile-like” than other similes? If so, what can be our basis for judging this? Are some effects of the similes more characteristic of the most “simile-like” similes, with other effects being more

2

By “Hosea,” we mean in this study, for purposes of convenience, both the prophet and the book bearing this name. The diachronic development of the book not being in focus for us, we shall not consider questions of redactions and multiple authorship. It is possible that these issues could shape to some extent the application of the model of simile as we develop it, but we have at present no principles to apply in this regard.

(16)

marginal? Are some combinations of concepts more likely to be be treated by similes than others? Also, are some discourse-level functions more likely than others to be filled by similes?

But we cannot examine a phenomenon by itself without taking into account contrastive phenomena. In particular, we shall take metaphor into account. In fact, this study was inspired by the thought, can we account for the distribution of similes vis-à-vis metaphors in Hosea?

1.2 Justifying the task

Much attention has been paid to understanding Biblical figures of speech, but the ability that has been developed to do linguistics within cognitive frameworks has only very recently been applied to the Biblical languages and documents. Here we must survey some of the needs for this application, which for our purposes shall be construed quite widely so as to embrace the concerns of both metaphor and simile.

Application of Cognitive Linguistics to Old Testament Hebrew has been scant. Mandelblit has shown the way in some respects, having analyzed the Modern Hebrew verb in construction grammar terms (Mandelblit 2000). As for semantics, Brettler (1989), Hermanson (1995, 1996), and Stienstra (1993) have identified and discussed various conceptual metaphors. There has been no attempt, so far as we know, to apply the mechanisms of conceptual blending to the Hebrew Bible in respect either to speech figures. Cognitive semanticists have recently begun paying attention to speech figures other than metaphors, and there are intriguing hypotheses about the relationship of metaphor to simile. These initiatives are too new to have been applied extensively to Old Testament Hebrew, but there are several reasons to attempt an application.

First, it is desirable to apply the insights of conceptual blending and metaphor typology to a non-living language. The cognitive semanticist has traditionally relied to a very great extent on native speaker intuition and idealized utterances. Investigation has normally proceeded on the basis of a well-known culture and a well-known language. But how would the application of these same insights fare in the case of a culture and a language very much less known? Is it possible at all to work backwards, from language to culture, in an attempt to recover cultural insights? This is, of course, the approach of Brettler, Hermanson, and Stienstra, and it must be part of our approach, also. But it is one thing to adduce, as they have done, certain Hebrew conceptual metaphors and their actual realizations in Biblical texts. It is another thing to apply conceptual blending insights to metaphors and similes, for this may require a considerably greater knowledge of the Biblical culture in general, knowledge which may be lacking.

(17)

Secondly, it is desirable to apply conceptual blending theory to the study of similes. The model and apparatus of conceptual blending were developed initially with reference to metaphor theory in a cognitive framework. One of the unlooked-for results of this application will be the demonstration that conceptual blending analysis becomes an indispensable tool for achieving a profound understanding of the simile in question.

Thirdly, if any kind of relation, whether one of content or function, between metaphor and simile can be identified, it should be done. This would be very desirable on a universal level, as well as in BH. If such relationships can be shown to be language-specific, they should be so shown. And one should then ask, what does this mean for Bible translation theory and practice?

Fourthly, in addressing the need to understand similes, it is desirable to come to terms with the theoretical claims of the model of metaphor developed by George Lakoff and others. It is not enough to adopt some pieces of the model, as Brettler, Hermanson, and Stienstra have done with the conceptual metaphor construct, without addressing the theoretical stance looming behind these pieces. Once that stance, however, is addressed and carefully considered, especially in the light of the pertinent disciplines of cognitive anthropology and cognitive psychology, we shall find good reason to seek out a less daring and more defensible approach to Biblical language and Biblical culture.

This volume actually wishes to contribute to a much larger goal than the previous points suggest: if we might crudely summarize those points by saying that we wish to “get inside the heads” of the ancient Hebrews, to see the world as they saw it, then it is not enough to learn their worldview(s) and accompanying concepts, as daunting a goal as this appears to be. We also desire to see how these concepts are mixed and manipulated; we want to work toward defining, so to speak, a grammar of conceptualization. What preferred ways for presenting and blending concepts does BH have? How many of these ways can we identify? Once identified, can these patterns of conceptualizations aid us in the age-old problems of Biblical textual criticism and the interpretation of BH lexemes and phrases?

In our stated purpose of this volume, then, to discover by examining the book of Hosea the principles of “Embodiment and Judgments of Prototypicality in Forms, Functions, and Conceptualizations,” we hope to contribute towards an over-arching goal of defining a BH grammar of conceptualization.

(18)

1.3 Taking measure of the opposition to this task

Opposition is always valuable: it keeps one from presuming to account for all data. Moreover, it compels one to reckon with the force of tradition and to weigh the insights of those who have gone before. Finally, opposition can drive the investigator to firmer ground for his own work than he otherwise might have bothered to search for. Let us look below at three kinds of opposition to various aspects of the task outlined in the previous section.

1.2.1 Opposition from the code model of communication

From linguists dedicated to the code model of communication, including a strong view that communication is mainly effected propositionally, would come opposition to a Cognitive Semantics framework. The modern Bible translation field, for example, appears to be dominated by a rather Lockian view that concrete language is the honest tool to use in communication and that, if figurative language need not always be condemned for “stirring the passions,” it is at least optional and may be dispensed with in translation. Certainly Beekman and Callow (1978), judging by their proposed solutions to translation problems posed by metaphor and simile, appear to be in agreement with the view that figures of speech tend to be incidental to the weighty, concrete substance of language. Their view of metaphor and simile3 as essentially comparisons between two domains ( “source” and “target”), their view that the “target’s” relevance to the metaphor may almost always be reduced to a single component of meaning, and their frequent willingness to dispense entirely with a speech figure in translation—all of these factors suggest a view that metaphor and simile are at best secondary in a mainly propositionally-driven model of communication.

1.2.2 Opposition from some quarters of anthropology

From many anthropologists might come opposition to the concept of worldview, or indeed, the construct upon which it depends, the concept of culture itself. The very concept of culture has lost much support in the discipline of anthropology during the past ten or fifteen years. Contributing factors seem to have included the realization that, if it exists, culture is not monolithic, but varies among every class of person; that history belongs to the strong and is normally written by the dominant; that ethnic representatives often relay as truth what they wish to be told; and that even if one relays what one believes, it may be contradicted by a more real state of affairs (Strauss and Quinn 1997:3).

From certain anthropologists might also come disdain for the topic of figures of speech. Ohnuki-Tierney (1991:160) identifies a strain of anthropology that viewed such a study as proper only for

3

(19)

students of great literature. This is in spite of the great tradition of trope studies4 in anthropology, some of which are mentioned in a convenient summary by Fernandez (1991:3–4); he cites as particularly formative in this regard the work of Claude Lévi-Strauss, including The Savage Mind, but also more recent work of Stanley Tambiah, James Fox, and Michelle Rosaldo.

1.2.3 Opposition from within Cognitive Semantics itself

From those following in the immediate tradition of the most prominent studies within a Cognitive Semantics framework,5 studies which have largely been restricted to treating metaphor, there might come surprise that other figures of speech might need to be treated in order to understand worldview. It is as if the immense strides taken in metaphor study of the recently opened area of Cognitive Linguistics have been taken so fast, that there has been little time to examine other figures of speech in the same light. One might cite as another reason for this omission the venerable tradition, extending back to Aristotle, that metaphor is the strongest of the speech figures, and that simile is its weaker sister (Soskice 1985:58).

1.4 Situating the task: cognition and embodiment

In this study, we shall approach Hosea’s similes from a cognitive perspective. A notion of cognition can have various aspects built in, e.g., prototypicality, entities irreducible to smaller parts, and embodiment. While we shall pay attention to all of these and more, we shall encourage the reader at this point: if one leaves this chapter with a single idea, let it be that embodiment will be the most crucial notion for our view of similes, as it will surface at strategic times in our discussion.

We understand embodiment as the ultimate grounding of concepts or their categories in human experience or ability. Lakoff (1987:12) writes:

[Conceptual embodiment is] the idea that properties of certain categories are a consequence of the nature of human biological capacities and of the experience of functioning in a physical and social environment. It is contrasted with the idea that concepts exist independent of the bodily nature of any thinking beings and independent of their experience.

4_{“Tropes” as used by Ohnuki-Tierney (1991:162) is a general term for several kinds of metaphor which have been of interest to} anthropologists: Ohnuki-Tierney lists as tropes metaphor (in a narrow sense), metonymy, synecdoche, and irony. Polytropy is, then, the phenomenon of cultural symbols functioning in more than one of these ways, either synchronically or diachronically. A point arising from Ohnuki-Tierney (1991:186) and filled with potential implications in the study of texts is that one can never with certainty fully distinguish a pure referential meaning of a lexical item from a poetic meaning, or, from, ultimately, even a cosmological meaning. It is also noteworthy that the phenomenon of polytropy challenges the traditional notion of metaphor as the “master” trope, i.e., “the most creative and powerful and all tropes” (Ohnuki-Tierney 1991:184).

5

We include in this category works that have become standard in the development of Cognitive Semantics and the Lakoff-Johnson model, e.g., Lakoff and Johnson (1980), Lakoff (1987), Johnson (1987), Lakoff and Johnson (1999).

(20)

We understand prototypical categorization, where categories are said to have central and noncentral members, as being linked to embodiment; embodiment also accounts for the fact that some experiences and their structures are given to us as humans ever before we adopt a conceptualization of them—the process which Lakoff (1987:302) calls “preconceptual structuring.”6 Moreover, embodiment appears to us to explain gradations in concreteness that characterize various linguistic phenomena, ranging from prototype effects among English nouns (see Section 2.1.5); to certain characteristics of our posited Major Simile Type (see Section 4.3.3), such as our term Imaged State of Being and the Major Similes’ tendency to project strong semantic properties to following text; and to the gradation of prototypicality that we shall hypothesize exists among simile types (see Section 4.9). In summary, the notion of embodiment shall repeatedly offer satisfying explanations in the course of this study.

1.5 Outlining this study

Chapter One (Introduction: The Task and its theoretical background) presents our task—that of accounting for Hosea’s similes from a Cognitive Linguistic standpoint; the chapter also presents some obstacles it faces from other areas of scholarship, as well as the broad notions on which it rests: embodiment and Prototype Theory.

In Chapter Two (A Cognitive Linguistic Approach to Similes), principles of this study are established within the realm of Cognitive Semantics and Cognitive Syntax. As for Cognitive Semantics, a weakened version of the Lakoff-Johnson conceptual metaphor theory (often referred to as the Contemporary Theory of Metaphor) is adopted. The notions of embodiment and judgments of prototypicality are then presented and discussed.

Basic elements of conceptual blending are also presented in Chapter Two. Conceptual blending is shown to be a very widespread process, applicable to syntax as well as to semantics. This observation is meant to prepare us to follow in detail (in Chapter Six) the blending of concepts in selected similes of Hosea.

From the normal application of the basic Lakoff-Johnson model and conceptual blending to conceptual metaphors and image metaphors, Chapter Two advances to the study of similes. We argue that, besides considering the semantic and conceptual structure of similes, one must also consider the essential differences between metaphor and simile on the basis of usage in text.

6

Lakoff (1987:302) suggests that preconceptual structure is found in basic-level categories and kinaesthetic image schemas (see Section 2.1.6.2).

(21)

Finally, we apply the theory and apparatus of conceptual blending to similes. The nature of simile requires us to add an apparatus accounting for the syntactic structure of similes.

Seeking to establish a basis to relate similes to culture, Chapter Three (A Cultural Approach to Similes) uses the cognitive anthropology of Strauss and Quinn in the area of cultural meaning to present several cultural constructs: cultural schemas, cultural exemplars, cultural models, and cultural themes. Strauss and Quinn’s conclusions about how metaphors are used in everyday speech are shown to be in accord with our postulation that speaker assessment of the hearer usually includes an assessment of the hearer’s ability to process language in general and, in particular, to process every utterance, usually before it is produced by the speaker. The necessity for this assessment by the speaker provides us with a text-based platform for erecting one part of a theory of simile. Another part of the theory concerns a generally-held view of relative strength of simile and metaphor: that metaphor is “stronger” than simile. We are able to qualify this view and give it precision in a way that turns out to be counter-intuitive, yet consistent with various data.

Chapter Three then presents a model for the integration of metaphor with a theory of meaning in culture, following the work of Strauss and Quinn, and proposes to integrate the Strauss-Quinn cultural meaning model with Boroditsky’s Weak Structuring view of metaphor. The effect is to provide a reasonable basis, amenable to empirical investigation, for the investigation of both metaphor and simile. Finally, Chapter Three applies the notions of embodiment and prototypicality to the cultural constructs of Strauss and Quinn.

Chapter Four (Toward a Prototypical view of Hosea’s Similes: model and methodology) presents a wide range of assumptions and conclusions that are later used to analyze Hosea’s similes. Chief among these are the following: the basics of Floor’s (2004a) model of Information Structure for Biblical Hebrew narrative, with modifications and additions for BH poetry—some from Floor and some from us (including the notion of macro frame), are adopted for this study; we hypothesize for working purposes three cognitive types of similes in Hosea; and we adapt the metaphoric conceptual blending apparatus to similes.

Perhaps the most striking impression that conceptual blending diagrams, introduced in Chapter Two, leave with the analyst is the portrayal of some of the immense conceptual complexity that often exists in metaphors and similes—complexity that is mastered by the human mind without very much awareness of doing it. This complexity proves, in our analysis, to be complemented by regular patterning among types of conceptual associations, which we call conceptual manipulations. It is this patterning, justified by principles of embodiment, that we hypothesize to account for the

(22)

much of the distribution of similes versus that of metaphors in BH poetry. We then find that we can integrate these patterns with the three simile types already established for working purposes; and we correlate the cultural constructs of cultural schema, cultural theme, and cultural model, drawn from Strauss and Quinn’s anthropology, with the notion of embodiment and with Hosea’s similes and metaphors.

Chapter Five (A Brief Survey of the Book of Hosea) briefly presents our assessment of the dating and political situation of Hosea; we also discuss issues concerning the textual integrity of the book, agreeing with a scholarly consensus that sees signs of sophisticated editing of the Hosean material. At the same time, we see many marks of the orality that we presume to be at the basis of Hos. 4–14. We come to the conclusion that, regardless of the editorial overlay or overlays that may be represented in Hosea as we know the book to be, we must, for our purposes, analyze the text as we have it, insofar as we can reasonably do so, and not attempt to strip away the redactor’s work. This is one principle that we use to characterize the principal authorities whom we cite in our examination of Hosea’s simile in Chapter Six.

Chapter Six (A cognitive examination of Hosea’s similes for prototypicality) deductively applies all the foregoing theory to an examination of Hosea’s similes. In this way, we show our theory of similes to be coherent. However, this examination of Hosea’s similes is also inductive in that it makes other observations leading to additional conclusions pertinent to our cognitive study of similes in Hosea: (1) kinaesthetic image schemas, one kind of cognitive linguistic construct, are shown to contribute to some similes’ conceptual blending and to be the basis of a number of poetic chiasms—something that we would not have predicted from our cognitive theorizing in Chapter Two; (2) the distribution of similes vis-à-vis metaphors for the purpose of image elaboration is systematic; and (3) similes participate in Information Structure, in that their Vehicles function like the more “standard” verbal arguments, since they may occur in marked or unmarked position. Chapter Seven (A Summary of Embodiment and Judgments of Prototypicality in Forms, Functions, and Conceptualizations in Hosea’s Similes) summarizes this study’s research and conclusions. Prominent among the conclusions is our final decision to “abolish” the three cognitive simile types that we had established for working purposes, having no further use for them; their help is no longer needed for us to recognize and account for the embodiment and judgments of prototypicality that characterize Hosea’s similes. A second prominent conclusion is that for Hosea and his audience, their view of YHWH, although proclaiming him as transcendent and wholly “other,” nevertheless depended in a profound way upon knowledge of themselves and their environment. A third major

(23)

conclusion is the effects of this study upon interpretation of the Book of Hosea and, by extension, on the rest of the Hebrew Bible.

Chapter Eight (Further Directions) discusses prospects for further research and possible implications for translating Hosea’s similes and metaphors. A study of this kind cannot help but create a “wish list” for the advancement of theory; here we present our wishes for a cognitive approach to Information Structure and for a cognitive approach to the distinction, held so intuitively by people, between literal and figurative language.

(24)

Chapter Two

A COGNITIVE LINGUISTIC APPROACH TO SIMILES

2.1 The framework of this chapter and the 1987 Lakoff-Johnson model 1

This chapter will develop an approach to metaphor and simile that is grounded in linguistics, raising certain issues which, in our judgement, cannot be resolved without an appeal to cultural anthropology, which we shall make in Chapter Three.

Cognitive literature discusses in general metaphor far more than simile or, indeed, any other trope. The linguistic approach laid in this chapter will often appear to discuss metaphor only, but we shall show that the principles developed early in this chapter can be applied to many linguistic structures besides metaphor. The fact that simile is not mentioned much at first is attributable mainly, in our view, to simile’s lack of treatment in general in the literature. Beginning in Section 2.6, however, we shall treat simile explicitly.

The general linguistic framework of this study will be Cognitive Linguistics, in the fields of both Syntax and Semantics. In contrast to formalistic approaches, this school of study sees language as part of general human cognition. It tends to be cross disciplinarian in nature, drawing in particular from every kind of study of human perception, and it consequently tends to project results and theories across disciplines as well. Anthropology, pedagogy, and theology are several fields which it has affected (Saeed 1998:300–302). We begin below the exposition of this approach by describing what we shall call the 1987 Lakoff-Johnson model.

The mature, working Cognitive Semantics model as developed over the years by Lakoff, Johnson, and others is usually taken to be the model presented in Lakoff (1987). The heart of this model’s apparatus is the Idealized Cognitive Model (ICM), which Lakoff (1987:68) terms “a complex structured whole, a gestalt.” The ICM corresponds to a great degree to what people would call a

1_{This model builds most immediately upon Lakoff and Johnson (1980), Johnson (1987), Fillmore (1982), Langacker (1986), and} Fauconnier (1985). Its philosophical ramifications are notably expounded in Lakoff and Johnson (1999). Although a number of scholars gave important and primary input to the model over some years, the names of George Lakoff and Mark Johnson are certainly the most closely associated with it, and it is for that reason that we call it the Lakoff-Johnson model. The appellation in common usage is “The Contemporary Theory of Metaphor” (CTM), drawn from the title of Lakoff (1993), but a term which some might find rather presumptuous.

(25)

commonsense view of reality, and as such is a folk theory (Lakoff 1987), or a cultural model (Quinn 1991). An ICM may very well be far from any scientifically held or attested theory.

2.1.1 Frames

Lakoff (1987) bases the notion of ICM partly upon the frame semantics of Charles Fillmore. Fillmore (e.g., Fillmore 1982) building upon work of John Austin (e.g., as in Austin 1961) claimed that, contrary to the impression given by dictionaries, word senses cannot be so neatly and succinctly defined, but that in reality they imply, and therefore depend upon, entire gestalts. For example, an account of the concept of a restaurant waiter cannot be complete without the presentation of the institution of restaurants, the habits and expectations of diners, the place of restaurants and their workers in the economy, etc. Each of these aspects of restaurants’ existence comprises a gestalt, or, in Fillmore’s terms, a frame. It is clear that frames are connected in various ways to other frames, and that some are embedded in others. Frame connectivity is in a sense matched by lexical connectivity: Fillmore pointed out that some words typically evoke identical frames, while focusing upon different aspects of the same frames. Thus, restaurant, waiter, chef, and diner refer to various aspects of one frame. Coulson (2001:18) gives a helpful summary statement about Fillmore’s frames:

Fillmore defines a frame as a system of categories whose structure is rooted in some motivating context. Words are defined with respect to a frame and perform a categorization that takes the frame for granted.

In cognitive models of language, the idea that word senses evoke frames instead of definitions accounts for the intuition that when someone hears even a “simple,” “literal” utterance, much more than a “simple” meaning is normally constructed. Instead, a large amount of background and often explanatory information is accessed as well.

It was M. Minsky (1975) who in 1975 proposed the notion of frames, thinking of them as familiar events, conditions, and situations. An important feature to the theory was slots and fillers, which were tightly associated together. Slots were expected occasions for elements and relations in frames, and fillers were the expected, typical elements and relations, etc. As an example, consider the frame of a criminal trial. Among the typical slots of this frame are certain officials, certain groups of people, certain roles to be filled, and certain spatial relationships. The fillers of these slots are judge, defendant, prosecuting and defense attorneys, the jury, spectators, the press reporters, and the bailiff. Roles to be filled in the roles of moderating and keeping order, prosecuting and defending the defendant, and coming to a verdict.

(26)

Much of the power of frames lies in their prototypicality. Each slot-filler combination has a default value; any failure to specify a value results in the presumption of the default value, and any deviation from the default value can be easily noted.

2.1.2 The idealized quality of Idealized Cognitive Models

Lakoff characterizes ICMs as idealized: they receive their structures from a small number of situations that are considered to be typical or most representative. Waiter, for example, evokes the notion of a restaurant with a waiter staff distinct from the cooking personnel. One assumes also an eating establishment with places to sit down. Indeed, the term restaurant itself is likely to be imagined in such a way as well. One is not likely to speak of waiters in the context of a fast-food restaurant, and, in America at least, the term fast-food restaurant is normally considered to be distinct from restaurant. The fast-food restaurant is a less prototypical restaurant than the establishment referenced by the term restaurant.

2.1.3 Mental spaces

The immediate function of ICMs is to organize people’s knowledge, beliefs, and impressions. They do so by structuring mental spaces (Fauconnier 1985), which are purely conceptual, existing only in the mind but not in reality, having no rapport to any theory linking symbols to the real world. ICMs have also a transcendent function: that of structuring and channeling much of the reasoning process, this being one of the ICM model’s most telling claims.

2.1.4 The structure of Idealized Cognitive Models

ICMs are built upon and are structured according to four kinds of principles: propositional structure; kinaesthetic image schemas; metaphoric mappings; and metonymic mappings. Abstract ICMs are usually structured according to kinaesthetic image schemas (see below).

2.1.5 Prototype effects and prototypical scenarios

ICMs are the source of prototype effects. In fact, ICMs represent prototypical scenarios (Lakoff 1987:68–74), which is one reason they are called idealized. ICMs vary as to how well they actually fit people’s knowledge or belief about the world. Lakoff points out, for example, that one would hesitate to call either the Pope or Tarzan of the Apes a bachelor, even though neither is married. This is because bachelor belongs to the ICM of unmarried men whose circumstances are normal enough to allow them to marry. The Pope has made himself ineligible, and Tarzan is unable to marry for reasons beyond his control, so neither really fits the bachelor ICM.

(27)

Prototype Theory stands in contrast to Set Theory, the classical theory of categorisation. In the latter, every defining feature of a category must, at least in theory, apply equally well to all category members. Category boundaries are thought of as sharp and distinct.

In Prototype Theory, category defining features apply to category members to various degrees, such that some members are considered to be more representative of their class than others, and some may be thought to be best representatives. Thus Lakoff writes of “asymmetries”—differences in the cognitive standing among various members of a category, which account for the prototype effects. These concern the notion of markedness. Thus, for example, English nouns in their singular number are said to be unmarked for the category of number, while plural nouns are said to be marked. The unmarked state has a cognitive significance; Lakoff (1987:59–60) comments:

The intuition…is that singular is, somehow, cognitively simpler than plural and that its cognitive simplicity is reflected in its shorter form…, in simplicity of form. Zero-marking for a morpheme is one kind of simplicity.

Lakoff (1987:60–67) adduces other examples in Linguistics of the link between the notions of markedness and cognitive simplicity. For example, in phonology, unmarked consonants are frequently considered to be easier to articulate than marked. In semantics, he remarks that a speaker, in asking How tall is Harry?, makes no implication about his height; whereas if he asks, How short is Harry?, he implies that Harry is not tall. Thus it is said that the potential contrast between tall and short for the feature of giving implications is neutralised; tall is unmarked for implications, and is cognitively simpler than short.

1 Parallel expressions 2 Passive expression 3 Gapping 4 Pluralisation

To stub one’s toe A stubbed toe I stubbed my toe, and she

hers.

They stubbed their toes. *They stubbed their toe.

To hold one’s breath *Held breath *I held my breath, and she

hers.

They held their breaths. They held their breath.

To lose one’s way *A lost way *I lost my way, and she hers. *They lost their ways.

They lost their way.

To take one’s time *Taken time *I took my time, and she

hers.

*They took their times. They took their time. Figure 2.1.5

(28)

We wish to pay more attention to a syntactic example, where Lakoff cites Ross (1972, 1973a, 1973b, 1974, 1981); Ross attributes to “normal” English nouns the full set of grammatical manipulations possible in English (e.g., pluralisation, use in passive sentences, and use in gapped sentences). Ross (1981) begins with the parallel expressions, displayed below in column 1 of Figure 2.1.5:

Lakoff calls toe “nounier” (i.e., more prototypical a noun) than breath, which in turn is “nounier” than way, etc. It is only the “nouniest” nouns that qualify for the full range of syntactic manipulation available in English. We would add to Lakoff’s analysis by observing here that the “nouniest” noun in this list also happens to be the cognitively simplest; that is, we judge toe to be the simplest and most concrete, breath to be the next simplest and concrete, and time to be the least simple and concrete. The correlation among concreteness, cognitive simplicity, and prototypicality lies at the heart of this present study and will loom very large when we hypothesize and describe in Section 4.4.3 our three cognitive types of BH similes and the “Imaged State of Being.”

2.1.6 The ontology of Idealized Cognitive Models and Prototype Theory

ICMs are said to be composed of entities, predicates, and events (Lakoff 1987:399–400). These elements are expressed by constitutive metaphors, e.g., “anger is an entity,” “anger is a force.” The semantic domains of these constitutive metaphors are usually highly abstract, superordinate level concepts.

2.1.6.1 Conceptual metaphors

ICMs are represented in everyday language by conceptual metaphors. These map the constitutive metaphors onto language, e.g., the constitutive metaphor “anger is an entity” accounts for the conceptual metaphor phraseology ANGER IS A…. The effect of this constitutive metaphor is that

people are convinced that anger exists as an entity. The constitutive metaphor “anger is a force” accounts for possibility, in turn, that the conceptual metaphors ANGER IS PRESSURE and ANGER IS A STRUGGLE can exist in English. Other conceptual metaphors of anger are ANGER IS A HOT FLUID,

ANGER IS A FIRE, ANGER IS INSANITY, and ANGER IS A BURDEN.2 It can therefore be said that, whereas constitutive metaphors express the ontology itself, conceptual metaphors and metonymies describe its functioning.

It should be noted that a conceptual metaphor is not a metaphor at all in the popular sense. It is instead extra-linguistic, having only a conceptual existence, actually representing “a mode of thought” (Lakoff 1993:210). Evidence for the existence of a conceptual metaphor is the ensemble of

2

(29)

metaphorical expressions which are licensed by it. Thus, for example, the conceptual metaphor

ANGER IS A FIRE licenses expressions such as Joe’s anger blazed up, Cool down!, Don’t get so hot

under the collar!, and His anger smoldered for days.

The motivation of a conceptual metaphor lies in its mapping of semantic structure from source to target domains. The mapping proceeds according to shared knowledge about both domains. The mapping is effected by embodied kinaesthetic image schemas (see Section 2.1.6.2).

One of the strongest claims about conceptual metaphors is that they tend to define the parameters and nature of our thinking about a certain subject and that, as a result, they tend to be the lenses through which we view and reason about that subject. They allow us to comprehend the subject and to make inferences about it. The importance of conceptual metaphors is illustrated by the fact that it can be very difficult to talk about a subject, e.g., anger, without using them. Lakoff has in fact been read by some as holding that all of the human reasoning process is directed by conceptual metaphors, but he himself has told cognitive anthropologist Naomi Quinn (Quinn 1991:59 footnote) that he never intended to be understood to that extreme. Lakoff would rather hold that metaphor partially constitutes understanding.

One common way in which a central conceptual metaphor is productive is to give rise to an elaboration of itself. Consider the expressions to stew and to simmer as metaphors for enduring anger. These cooking terms have been borrowed to express the central conceptual metaphor ANGER

IS THE HEAT OF FLUID IN A CONTAINER (Lakoff 1987:384).

2.1.6.2 Embodiment and Prototype Theory

Lakoff and Johnson claim that constitutive metaphors and conceptual metaphors are grounded in human interaction with oneself and one’s environment. Thus human reason, which can be very abstract indeed, hangs upon this interaction. The tighter the links between any segment of reasoning and human physical interaction with oneself and the environment, the more embodied that reasoning is said to be. Human physical interaction with oneself and the environment is, at heart: (a) preconceptual in nature: we began to engage in this interaction even before being born, and it continues for the most part to be “aconceptual,” since it includes physical sensations and motor activities, generating neuron pathways in the brain; and is

(b) structured in a way that allows us to generalize it and treat it abstractly in our minds. In this interaction we recognize two kinds of structures: basic-level structures and kinaesthetic image schemas.

(30)

We shall find that embodiment is the golden thread that runs through our analysis of Hosea’s similes. It is the theory of embodiment that accounts for Prototype Theory and that lies at the heart of this volume.

Following the work of Eleanor Rosch (e.g., Rosch 1973, 1977), Lakoff and Johnson adduce the following levels of a taxonomic hierarchy characteristic of Prototype Theory, with examples from Lakoff (1987:46) added:

Superordinate level animal furniture Basic Level dog chair Subordinate Level retriever rocker

Figure 2.1.6.2

Taxonomic levels characteristic of Prototype Theory

The basic level is termed thus for many reasons: it is the highest level on which its members are perceived of as irreducible gestalts and as possessing similar shapes (e.g., all dogs have crucially similar shapes); it is the level learned by children first (e.g., children learn dog before they learn retriever); and it is the highest level whose members call for nearly identical interaction from humans (e.g., we interact with nearly all dogs in a similar way). In summary, basic-level categories merit being called “basic” by virtue of their perception, function, communication, and organization in human knowledge. Because of all the characteristics of basic-level phenomena listed above, basic-level concepts can be said to be rooted in physical experience.

To finish describing the taxonomic hierarchy, let us note that the superordinate level tends toward abstractness. This is the level on which most conceptual metaphors seem to exist (Lakoff 1993:211). The subordinate level tends toward the technical, often including elements that are rather unknown or even unfamiliar to people. Some might find it difficult, for example, to distinguish a beagle from a terrier.

Thus, we claim that of these levels, it is the basic level that displays the most embodiment, i.e., the closest links with human physical interaction with oneself and the environment.

There are what we shall call subconceptual elements as well, phenomena that can be viewed as contributing to the structure of a discrete concept. Johnson (1987:29) describes a kinaesthetic image schema as “a recurrent pattern, shape, and regularity in, or of, these ongoing ordering activities,” such activities being “our actions, perceptions, and conceptions.” It is these schemas which give motor coherence to our body’s spatial movements and to our manipulative and ultimately perceptual interaction with ourselves and our environment. Image schemas have a very

(31)

small number of parts and relations among them. Examples are the FRONT-BACK schema, the

LINEAR ORDER schema, the CONTAINER-CONTAINED schema, the IN-OUT schema, the FORCE

schema, the CENTER-PERIPHERY schema, and the BALANCE schema. We begin to perceive reality

and to interact with it in terms of these schemas presumably even before leaving our mother’s womb. Thus, for example, we learn that things can contain other things; we then learn various entailments of this principle, e.g., that containment can be embedded within containment, and that removal of an object from the embedded containment does not necessarily entail its removal from all containment. We learn that force can be applied against force, or against inert stationary objects. We learn that force can be augmented, diminished, or interrupted. With these elementary lessons is laid much of the foundation of our perceptual and reasoning capacity.

Lakoff (1987:444–145) makes the point that “image” in this context is not necessarily to be taken literally as always involving the faculty of sight. Kinaesthetic image schemas such as SOFT-LOUD

and FORCE may only obliquely involve sight. Certainly sensorily-handicapped persons may move in

and perform manipulations with the environment often as well as others. This leads Lakoff (1987:446) to conclude that

much of mental imagery is kinesthetic—that is, it is independent of sensory modality and concerns awareness of many aspects of functioning in space: orientation, motion, balance, shape judgments, etc.

Kinaesthetic image schemas have a certain, fundamental logic, reflecting their utterly embodied nature. They all:

(a) have structure, e.g., the CONTAINER schema has a boundary, which separates an interior from an exterior. The body is viewed as a container, but so are many other items, as well as activities and abstract concepts, these last of which make even more metaphorical the entire process (Lakoff 1987:271), e.g., one comes out of a stupor (Lakoff 1987:272).

(b) are gestalts, cognitive wholes, whose parts make no sense without the whole. (c) are considered meaningful by people because of their own physical experience.

(d) enable us to assign structures to perceived objects and events, thus allowing meaning to arise (Johnson 1987:29). When this logic is extended to other domains, including abstract domains, a metaphor results.

A cognitive analysis of similes in the book of Hosea